Hierarchical Hybrid Protocols for Wireless Sensor Network: A Survey

Hierarchical Hybrid Protocols for Wireless Sensor Network: A Survey

Abstract: Wireless Sensor Network, the name that explodes lot of crucial improvements in growing world technological impact. In WSN, clustering makes the vital role for the deployment of coverage, energy efficient, high localized communication between sensor nodes and its base station. To make the cluster process more effective various protocols are introduced especially in hierarchical protocol like LEACH, SEP, TEEN, PEGASIS protocols. In present scenario, the protocols superior process characteristics are combined that produces the Hybrid Protocol for various purposes like energy conservation, prolonged network life time, improved localized process, reduced no of CH selection. Here our work that made the survey of these hybrid protocols and their superior characteristics on the basis of no of nodes alive for defined round, stability period, energy dissipation under various conditions. Each protocol made the dominant characteristics in various aspects. Their advantages and their characteristics are discussed with their characteristics.

Keywords: LEACH, SEP, TEEN, PEGASIS, localization, Energy dissipation.

I.INTRODUCTION

Wireless Sensor Network is a stream where it composed of wireless sensor nodes deployed in a region to sense various type of physical quantity information from the environment. The information gathered by these sensors is then processed and sent it to the Base Station using Sink architecture. WSN plays a enormous role for various applications like military surveillances, industrial environmental control, Green-orbs monitoring, habitat monitoring, transport controlling etc. The sensor nodes are wireless and nano-size level which can be deployed with ease in remote areas and hilly terrains. Even though they are compact, their dense properties make them difficult in monitoring of those. Particularly in the region, where the physical intervene of the human being is not possible. Once the network established, their foremost task is sensing the defined physical quantity of information, and processing it then send it to the base station. Whenever the sensor nodes performs their mission, they have to spend their given energy for sensing those defined quantities and spend some more energy for transmitting those information to the Base Station.

When we are considering about the sensor nodes, they have some problem in the sense of physical parameters. The one is Ad-hoc deployment, nothing but the deployed regions of sensor nodes where it can be tossed by the aeroplane. That means there is no intervention of human-beings. It is one of the most due of this sensor nodes, where it identify its

connectivity and distribution in the irregular infrastructure regions. The second one is they have the responsibility for reconfiguration incase of any environmental changes. The third one is unethered property of the sensor nodes. The last one is dynamic changes, where the system must be adaptable to changing connectivity as well as modifying environmental stimuli. Before going to analyze the protocol, we have to look about the inner subsystem of the sensor node. It has four subsystems, one is computing subsystem where it consists of microprocessor that has the responsible for control of communication protocol and operates under various operating modes for power consumption. Second one is communication subsystem that takes the task of communication with neighbouring nodes with the four modes of transmit, receive, idle and sleep modes. Another subsystem deals with sensing subsystem that has the main factor of energy dissipation factor. The last thing is power supply subsystem that makes the heart of all system and also the leading factor to energy conservation process. Many routing protocols are already proposed to improve the performance of WSN. The routing protocol can be classified into flat, hierarchical, location based network routing protocols. In flat routing protocol, the sensor nodes are assigned with equal task and equal role. SPIN (Sensor Protocol for Information via Negotiation) and DD (Directed Diffusion) are considered in this network. In Hierarchical routing, the network is partitioned into clusters to achieve efficiency. LEACH, TEEN, SEP are the main members of this network. The other one, Location-based routing, the position of the node is the key factor for the determination of its optimal routing path. When compared with flat routing and Location-based routing protocols in WSN, clustering routing protocols have enormous advantages, such as scalability, data aggregation/fusion process, less load, less energy, more robustness, collision avoidance, latency reduction, load balancing and fault tolerance etc.

1. HIERARHICAL PROTOCOLS

   Before going to analyze the hybrid protocol of various network we have to analyze the basic types of protocol, their advantages and disadvantages too. Thus we can get the thing of purpose for which hybrid brings the advantages and which hybrid make the contribution to improve the network in terms of coverage, no of nodes alive in extended round of communication, and also the prolonged lifetime of the network. Here in this section we have to analyze the protocols of LEACH, TEEN, PEGASIS, SEP. This protocol made the combinations which produce the

   hybrid protocols to improve the performance

   in various

   When CH receives all the messages from the nodes that are in

   aspects where we have to see that in the next section.

   the group, based on no of nodes in the cluster the CH creates a TDMA schedule and assigns each node with time slot when it

   A.LEACH

   LEACH means Low Energy Adaptive Clustering Hierarchy. It is the first hierarchical routing protocol in WSN. In this, there are two types of nodes are classified. One is normal sensor node and another one is Cluster Head (CH).

   Initially the normal nodes are formed and made the cluster

   can transmit. Leach protocol is a complete distributed network, where it requires the global information about the network. There are lot of changes that has occurred in this protocol, and made several modified advantages too. The most predominant advantages of leach is (i) Any node that act as CH in certain round cant be selected as a CH again. So each node shared the load imposed. (ii) It prevents unnecessary

   group. Then among all, one of the nodes is elected as CH. The selection process is based on the randomized procedure with assigned random value. The randomized value decides the CH with the comparison of threshold value T(n). When the random value is less than the T (n), the corresponding node

   leads the group as a CH. The threshold value is calculated by,

   collisions. (iii) The cluster can open or close the communication interface that avoid excessive energy dissipation with their allotted time slots. However it has some disadvantages (i) Leach performs the single hop inter cluster network communication which leads to direct routing between CH to BS, thus it was made as the inability network in large-

   region networks.(ii) CH selection is based in terms f

   probabilities not in energy considerations. May be the fact, the

   low energy node tend to act as a CH which holds lot of

   Where n is the number of nodes, p is the priori probability of a node to be elected as CH, r is the random number between 0 and 1 that is selected by a sensor node. If it is less than the threshold level, then the node react as CH. G portrays the set of nodes that has not elected as Cluster Head. In each round,

   the process begins with a set-up phase, the time where the

   problems and in coverage.(iii) The dynamic clustering brings more overhead.

   B. TEEN

   Threshold sensitive Energy Efficient sensor Network protocol (TEEN) is a hierarchical protocol whose main aim is to cope with the sudden changes in the sensed attributes nothing but the physical quantity of the environment like temperature, pressure. It combines the hierarchical technique in line with a

   clusters are organized followed by a steady state phase;

   data-centric approach. In TEEN, a 2-tier clustering topology is

   several frames of data are transferred from the nodes to CH and CH to base station.

   used with two level of threshold is proposed. One is hard threshold and soft threshold is another one. The threshold

   portrays about to the sensing value must switch on its

   Fig.1 Leach Protocol Phases

   After the defined 1/p rounds, all nodes are again eligible to get CH. In LEACH, the optimal no of CH is estimated with 5% of the total no of nodes. The process made with the phase, which at current node broadcasts an advertisement message to the rest of nodes. The remaining nodes that received the advertisement message, decide the cluster group which it belongs to this round.

   Fig. 2 Leach Protocol Cluster Formation

   transmitter and have to make the report with CH. The softer threshold in the sense, that trigger the node to switch on its transmitter and transmit.

   Fig. 3 Teen Protocol Cluster Formation

   There is some task done by the hard and soft threshold values. The hard threshold tries to reduce data communications by making the nodes to transmit the data only when the physical attributes are with the range of interest. The soft threshold further reduces the communication latency, where it processed

   the task only by little or no change in the sensed attribute. Using those threshold values, users can make the trade off between energy efficiency and data accuracy by the parameter adjustment. If we discussed the advantages of TEEN, one is controlled data processed is done with the threshold values.

   transmission at the end of the chain is processed by the token passing approach. In fig 4, the data from c0 is transmitted to c2 through c1. When the data is received by c2 it passes the token to c4 to transmit the data. Thus it coverage the total transmit process through token passing approach. When we go

   (ii) It is applicable to the reactive scenes and environmental applications. But there is

   time critical also some

   for the advantages of PEGASIS, (i) It reduces the problem of overhead arises in the dynamic cluster formation which is

   disadvantages too. (i) It is not suitable for periodic reports applications (ii) There may be wasted time slot and the BS may not be able to distinguish dead nodes from alive ones.(iii) If CHs are not in the communication range, then the data will be lost.

   main drawback of LEACH. (ii) The energy load is uniformly distributed in the network. However the drawbacks are (i) It is not suitable for time varying topology network. (ii) The long range communications between node to sink consume lot of energy. (iii) Excessive delay caused by single chain and high

   C.PEGASIS

   probability of bottle neck made this

   protocol into more

   Power Efficient Gathering in Sensor

   Information

   suffered. (iv) It is not very scalable because it if difficult to

   maintain the complete database about the location of all other

   Systems (PEGASIS) is an improved LEACH protocol. The concept is for each node to only communicate with their close neighbours and take turns being the head for transmission to

   nodes in the network.

   D.SEP

   the sink. In this protocol, the locations of nodes are

   randomized and have the ability of data detection, data fusion and positioning. And also energy loads are distributed evenly

   Stable Election protocol for clustered heterogeneous WSN is developed for two level heterogeneous networks. It

   among the sensor nodes in the network.

   includes two types of nodes, normal

   nodes and advanced

   Fig. 4 PEGASIS protocol cluster formation

   The basic thing of communication is done by the formation of chain which can be either assigned by sink and broadcast to remaining nodes or done by the nodes themselves using Greedy algorithm. When the chain is formed by them, the first thing is they tend to get the location data of all nodes and compute the chain locally using the greedy algorithm. In the process of chain formation, the assumption is made that all nodes have the global knowledge of the network. The chain construction is formed from the furthest node from the sink and the nearest neighbour which is closer to the sink will be the next node on the chain. When the node on the chain dies, the chain again reformed to bypass the dead node by the same manner.

   Data gathering process is done through each round by chain formation, where each node receives the data from one neighbour, fuses the data with its own, then transmits to its neighbour node on the chain. Thus the data is moved from one to another node at a random position on the chain. The data

   nodes based on the initial energy. The probability of threshold which each node Si uses to determine the chance to become a cluster head in each round,

   Also for two level heterogeneous networks, Pi is defined as:

   is the optimal probability of each node to become CH.

   Fig. 5 Throughput of the Network for SEP and LEACH

   Clustering Routing protocols		LEACH	TEEN	SEP	PEGASIS
   Cluster Characteristics	Variability of cluster count	variable	Fixed	variable	variable
   	Uniformity of cluster sizes	Even	Even	heterogen eity	Even
   	Intra-cluster routing	Single-hop	Single-hop	Multi- hop	Multi-hop
   	Inter-cluster routing	Single-hop	multi-hop	Multi- hop	Single-hop
   Clustering process	Control Manners	Distributed	Distributed	Distribu ted	Distributed
   	Execution nature	Probabilistic	Probabilistic	Probabil istic	Probabilistic
   	Convergence time	Constant	Constant	Constant	Constant
   	Parameter for CH selection	Adaptive	Adaptive	Adaptive	Adaptive
   	Objective	Load balancing	Reactive scenes lifetime extension	Proactive reactive	Load balancing

   Clustering Routing protocols		LEACH	TEEN	SEP	PEGASIS
   Cluster Characteristics	Variability of cluster count	variable	Fixed	variable	variable
   	Uniformity of cluster sizes	Even	Even	heterogen eity	Even
   	Intra-cluster routing	Single-hop	Single-hop	Multi- hop	Multi-hop
   	Inter-cluster routing	Single-hop	multi-hop	Multi- hop	Single-hop
   Clustering process	Control Manners	Distributed	Distributed	Distribu ted	Distributed
   	Execution nature	Probabilistic	Probabilistic	Probabil istic	Probabilistic
   	Convergence time	Constant	Constant	Constant	Constant
   	Parameter for CH selection	Adaptive	Adaptive	Adaptive	Adaptive
   	Objective	Load balancing	Reactive scenes lifetime extension	Proactive reactive	Load balancing

   TABLE 1 COMPARISON OF HIERARCHICAL PROTOCOL

   Attributes- set of physical parameter

   that is interested in

   obtaining of data. (ii) Thresholds- consists of hard and soft threshold. (iii) Schedule- this portrays the modified TDMA schedule. (iv) Count Time (CT)- maximum time period for the successive report made by a node.

   The superior feature of the APTEEN protocol is to switch between proactive and reactive modes to transmit data. All of the nodes in the network sense the environmental physical attributes continuously, but the nodes which sense the data value more than the threshold value permit the transmitting. If the node failed to provide the data for a time period which is equal to count time, it must sense the attributes and transmit the data again. In APTEEN, CH aggregates and transmits the data within its clusters. On the process of data aggregation, the data must be sufficiently correlated. The modified TDMA schedule makes the nodes task with time slot for transmission.

   When we considering the advantages (i) It combines both

   CHs election in SEP is done as randomized on the basis of probabilistic value which is presented in LEACH. By increasing m or , we can improve our system further. So SEP gives increased stability period and prolonged life time

   proactive policies which is in LEACH and reactive policies which is in TEEN. (ii) It employs a lot of flexibility with setting of CT interval and threshold value of energy consumption. It is not easy to overcome all the things here is also some disadvantages also there (i) Introduced additional complexity in assigning the threshold values (ii) The complexity process arise when implementing threshold based function correlated with attribute based naming of queries.

   B. CCBRP

   due to advanced nodes but at the same time two level

   heterogeneity made the increased throughput. Fig 5 portrays about the throughput rate of LEACH and SEP at m=0.2 and n=3. It clearly shows that in the basis of throughput SEP made a lag of form against LEACH protocol.

2. HYBRID PROTOCOLS

   So far we have seen the basic network protocol of hierarchical network. In order to overcome the disadvantages in the previous protocols in one with another, people produce their study work with combined effect of these protocols. They are nothing but the hybrid protocols. Here some of the hybrid protocols which make the high impact on the energy efficient with prolonged life time of network and also good coverage too.

   A. APTEEN

   Adaptive Threshold sensitive Energy Efficient sensor

   The goal of Chain-Chain Based Routing Protocol (CCBRP) is to achieve both minimum energy consumption and minimum delay. It divides the WSN into a no of chains with Greedy algorithm to construct the chain. The no of sensor nodes in WSN decides the no of sensor chains and no of sensor nodes in the network. Consider an WSN with N sensor nodes distributed in a 2D with size of L(m)x L(m). If number of nodes is hundred then assume the chain counts are equal to ten percent of N.

   Network protocol (APTEEN) is a extension to TEEN and

   make the goal of transmission of periodic data and reacting to critical events. It is based on the query system which employs 3 types of queries: historical, on-time, and persistent. And also QoS requirements are introduced in the process of on-time queries and overhead latency is reduced by the Modified TDMA schedule with a special time slot assignment manner. In this protocol, the CH broadcast four parameters (i)

   Fig. 6 CCBRP cluster formation

   In fig.6, the CCBRP cluster chain formation is shown where the nodes are grouped into 10 chains. In that C0 is the one of

   the node of initial chain and L5 is the chain leader of chain 5 and the ML is the main leader where it sends the data to BS.

   Then here, the secondary CH nothing but Master Cluster Head (MCH) are formed. After the formation of MCH, the normal

   The CCBRP has the two phase of process. In

   first phase,

   CH search for nearest MCH by calculating the distance

   randomly CH is selected in each chain. After that at the two ends, each node sends its data to its nearest neighbour node, the neighbour node fuse the data and send it to further nearest node. The processes are continued, till the leader receives the data and send the token to the end nodes. In second phase, after the all leader nodes from the various chains received the data then chain leader form new chain by Greedy algorithm.

   between them and transmit their aggregate data to the MCH. Then MCH receives the data and made the aggregation process and send to the BS. The population of CH and MCH is decided by predetermined fractional value ( p election probability) for CH and popt value for MCH.

   Then the node simultaneously sends the data neighbour. This process is continued until the

   its nearest Main leader

   receives the data. Main leader in the sense, it has the ability to send the data to BS. The average energy consumption is calculated by,

   The main proposal of this protocol leads to the 60% of less energy consumption than LEACH protocol and 10% of less energy consumption than CCM protocol. When in the sense

   Fig. 8 EEE LEACH network in Matlab

   In fig.8, it shows the matlab work and it compares with

   delay, CCBRP gives 75% less delay than PEGASIS which is

   LEACH and DTx process where it

   gives about 35% of

   same delay of LEACH protocol. When we

   go for the

   increased transmission time with 18% of reduced throughput

   minimum delay with minimum energy consumption level

   rate. In this work, the focus of improved is on reducing the

   network, CCBRP makes the good use of that. But the problem will arise when we go for larger area network with the bottleneck in the token passing and receiving the data from the neighbour nodes.

   throughput rate thereby increasing the process.

   D. TSEP

   energy conservation

   C. EEE LEACH

   Energy Efficient Extended LEACH protocol is the further step of LEACH where it focused with multi-hop

   Threshold sensitive Stable Election Protocol (TSEP) has made with two features. One is , it is reactive routing protocol and second is three level of heterogeneity. Thereby it

   begins with three types of nodes.

   process. The main aim of this protocol is to increase the

   energy efficiency by reducing the distance

   for radio

   (i) Normal Nodes

   communication. It involves the modification of LEACH by (ii) Intermediate Nodes

   introducing multi level cluster formation. First

   the normal

   (iii) Advanced Nodes

   cluster formation is done and the process the data with normal CH to transmit its own data.

   Advanced nodes had energy greater than all nodes, where the intermediate has the energy between normal and advanced nodes. Intermediate node is chosen by the factor b, fraction of nodes which have the relation that energy of normal nodes µ times more than normal nodes. In SEP, the normal nodes have the energy value of E0, for advanced node it has

   Eadv=E0(1+) and intermediate nodes have Eint= E0(1+ µ)

   where µ=/2. So the total energy of nodes will be

   where n is number of nodes, m is proportion of Advanced node to total no of nodes n and b is the proportion

   of intermediate nodes. The probability of node elected as

   normal, intermediate and Advanced through optimal probability is portrayed as follows.

   Fig. 7 EEE LEACH protocol cluster formation

   (Normal nodes)

   ÊÔ aú (Intermediate nodes)

   Êa aú (Advanced nodes)

   And at the same time the threshold value of these nodes are separated one. These node values are calculated by follow methods.

   Ê ú

3. CONCLUSION

The survey portrayed the clear view of hierarchical protocols and shows what are the modifications done in this and how the hybrid protocols are made. And also the advantages over other protocols are discussed with some experimental results. One of the major concerns to be notified is, we have to analyze the use of protocol. It is nothing but the area, where we have to use like whether we have to reduce the throughput

ú

1 Ê ú

d 1

Ê

Ê

ú

a ú

or whether we have to reduce the energy conservation on sensing process or with minimum delay with less energy

0 cdvanc

Ê

consumption or non-periodical data transmission. These

hybrid protocols are based on these approaches, mainly focused to increase the prolonged life time of the network

Ô

Ô

Ê

Ê

1 ÊÔ d 1

Ô

a Ô

with the trade off between the coverage and throughput.

ABOUT AUTHORS

0 cdvanc

He got UG degree from PSNA college of

Êa a

Engineering, Dindigul, Tamilnadu. He got PG

a 1 Êa

d 1

Ê

Ê

a

a degree from the same college. He attended

more conferences and presented more reputed

0 cdvanc

Where , , are the set of normal nodes, intermediate nodes and advanced nodes respectively. Even though the average no of nodes are equal in LEACH, SEP, ESEP the good aspect of TSEP is, it leads to reduced energy dissipation due to its heterogeneity node formation. Before going for the CH selection process, this protocol broadcast the four parameters.

(i) Report Time(TR)- the period which reports are being sent by nodes successively.(ii) Attributes(A)- the physical environmental parameter to be sent.(iii) Hard threshold- the value which the node will transmit the data to CH.(iv) Soft threshold- the value to switch on their transmitter and transmit in each node.

The most attractive features of TSEP are, the time critical data reaches almost instantaneously. Node sensing is done continuously but transmission is not done in frequent manner so energy consumption is lesser than proactive networks. The user can change the attributes of network depending upon the requirements. By the parameter like stability period, number of alive nodes per round and throughput, the proposal made the superior characteristics among LEACH, SEP, TEEN, ESE

Fig. 9 Comparison of TSEP with other protocol in terms of no of alive nodes.

papers in the field of Power Electronics and Drives. Currently he is working as an Assistant Professor in M.Kumarasamy College of Engineering, Karur. His interest is in the field of Power electronics control and Wireless technologies.

He got UG degree from Mepco Schlenk Engineering College, Sivakasi, Tamilnadu. He got PG degree from Anna Universtiy, Madurai. He presented more national and International Conference study papers and

published reputed journals. Currently he is working as an Assistant Professor in M.Kumarasamy College of Engineering, Karur His interest is in the field of VLSI based Power System Operation, Wireless Protocols, Image Processing.

He got UG degree from JJ college of Engineering and Technology, Trichy, Tamilnadu. He got PG degree from Sudharsan Engineering College. He presented many international and national Conferences in

many aspects. Currently he is working as an Assistant Professor in M.Kumarasamy College of Engineering, Karur His interest is in the field of Power Electronics and Wireless protocol technologies.